居家醫療介入對照顧者醫療資源使用的影響：照顧者年資效果分析

Abstract

背景：在人口老化的趨勢下，提升醫療服務的多元性與可近性以滿足各式醫療需求已成為當前重要政策議題。居家醫療作為以病患為中心的照護模式，不僅仰賴醫療團隊，也高度依賴照顧者的投入。而居家醫療除了改善病人的生活品質、提高照護連續性外，其對照顧者的影響亦極為重要。

目的：本研究旨在探討居家醫療介入是否會對照顧者醫療資源使用量產生影響，並分析此效果是否因照顧年資不同而有所差異。

方法：本研究串聯HOLISTIC (HOme-based Longitudinal Investigation of the multidiSciplinary Team Integrated Care)資料及全民健康保險研究資料庫，採前後期比較設計，分析被照顧者接受居家醫療前後一年內，照顧者醫療資源使用之變化（包含門診、急診、住院、精神相關用藥）。研究以多變項迴歸分析評估介入效果，並納入照顧年資及介入前後之交互作用項，以檢驗介入效果在不同照顧年資族群之間之差異。本研究亦進行敏感度分析與不同關係非正式照顧者次群體分析。

結果：本研究共納入308位照顧者。結果顯示，居家醫療介入後，整體照顧者之急診使用機率顯著減少10.5個百分點(95%CI: -0.204, -0.005)；非正式照顧者之住院點數於介入後顯著減少13403點(95%CI: -25971, -837)；配偶與同居人非正式照顧者之住院使用機率於介入後顯著增加20個百分點(95%CI: 0.02, 0.38)；其他關係非正式照顧者之住院點數於介入後顯著減少21581點(95%CI: -37422, -5740)、總點數於介入後顯著減少28596點(95%CI: -53437, -3756)。照顧年資與介入後之交互作用項，於所有分析均未達統計顯著，顯示對不同照顧年資之照顧者，居家醫療對醫療資源使用之影響效果並無顯著差異。此外，各群體介入前，針對部分醫療資源使用的分析顯示，照顧年資1~4年者，其介入前一年內醫療資源使用量，顯著少於年資未滿1年組，可能反映照顧歷程中的適應現象，或顯示在照顧者年資累積滿1年前，健康狀態不佳者就已退出照顧者群體。

結論：整體而言，居家醫療介入效果不因照顧者年資而有顯著差異，顯示居家醫療可能有助於照顧者緩解無法隨照顧年資累積克服之困難。然而，居家醫療之影響效果會因照顧者與被照顧者關係不同而有所差異。雖然效果大致呈現有助於減少醫療資源使用量之結果，然而對配偶與同居人非正式照顧者，介入後可能產生較高的住院機率，顯示居家醫療對配偶與同居人非正式照顧者之效果不同於其他群體，且可能存在負面影響。

Background: With the trend of population aging, enhancing the diversity and accessibility of healthcare services to meet various medical needs has become an important policy issue. Home health care, as a patient-centered model of care, relies not only on healthcare teams but also heavily on caregivers’ involvement. In addition to improving patients’ quality of life and promoting continuity of care, home health care also has a significant impact on caregivers.

Objectives: This study aimed to investigate whether home health care intervention affect caregivers’ healthcare resource utilization and whether these effects differ according to caregiving duration.

Methods: This study linked data from the HOLISTIC (HOme-based Longitudinal Investigation of the multidiSciplinary Team Integrated Care) and the National Health Insurance Research Database. A pre–post comparison design was adopted to analyze changes in caregivers’ healthcare utilization within one year before and after care recipients received home health care services, including outpatient utilization, emergency department utilization, inpatient utilization, and psychiatric medication use. Multivariable regression analyses were conducted to evaluate the intervention effects. Interaction terms between caregiving duration and pre–post intervention period were included to examine whether intervention effects varied across different caregiving durations. Sensitivity analyses and subgroup analyses among informal caregivers with different relationships to care recipients were also performed.

Results: A total of 308 caregivers were included in the study. The results showed that, after the home health care intervention, the overall probability of emergency department utilization among caregivers decreased significantly by 10.5 percentage points (95% CI: −0.204, −0.005). Among informal caregivers, inpatient expenditures significantly decreased by 13,403 reimbursement points(95%CI: -25971, -837) after the intervention. Among spouse and cohabiting informal caregivers, the probability of hospitalization significantly increased by 20 percentage points(95%CI: 0.02, 0.38) after the intervention. Among informal caregivers with other relationships to care recipients, hospitalization expenditures significantly decreased by 21,581 reimbursement points (95%CI: -37422, -5740), total healthcare expenditures significantly decreased by 28,596 reimbursement points (95%CI: -53437, -3756) after the intervention. The interaction between caregiving duration and the post-intervention period was not statistically significant in any of the analyses, indicating that the effects of home health care on healthcare utilization did not differ significantly across caregivers with different caregiving durations. Furthermore, pre-intervention analysis of healthcare resource utilization across groups showed that healthcare utilization among caregivers with 1 to 4 years of caregiving duration was significantly less within the year prior to the intervention than that of the group with less than 1 year of caregiving duration, which may reflect the adaptation phenomenon in the caregiving process, or show that caregivers with poor health had already left the caregiver group before their accumulated duration of caregiving reached 1 year.

Conclusions: Overall, the effects of home health care intervention did not significantly differ by caregiving duration, suggesting that home health care may help caregivers alleviate difficulties that cannot otherwise be overcome solely through accumulated caregiving experience. However, the impact of home health care varied according to the relationship between the caregiver and the care recipient. Although the overall pattern generally indicated a reduction in healthcare resource utilization, informal caregivers who are spouses or cohabiting family members showed a higher probability of hospitalization after the intervention. This finding suggests that the effects of home health care on this group may differ from those on other caregiver groups and may potentially involve negative impacts.